Project Summary/Abstract Alzheimer[unreadable]s disease (AD) is characterized by the deposition of amyloid B&#61472;protein (AB), a small peptide derived from B- and y-secretase cleavages of the amyloid precursor protein (APP). We recently demonstrated that the last 37 amino acids (LRP-C37) of low-density lipoprotein receptor-related protein (LRP) without the NPXY motifs to be necessary and sufficient to increase AB&#61472;production. Since LRP-C37 alone was a potent inducer of AB&#61472;production, we used this domain as bait in a yeast 2-hybrid screen, resulting in the identification of Ranbinding protein M (RanBP9). Indeed transient transfections of APP and RanBP9-FL or FL-derived RanBP9- N60 robustly increased secretion of AB&#61472;in varieties of cell lines, indicating that RanBP9 alters APP metabolism. Most importantly, immunoblot quantification of RanBP9 protein levels demonstrated that RanBP9-N60 and RanBP9-FL were elevated more than six and four-folds in the brains of AD patients and APP J20 transgenic mice respectively. We also found that like LRP and two of its key ligands, RanBP9 is genetically associated with late-onset AD. To gain a better insight on the in vivo role of RanBP9 in the pathogenesis of AD, we generated transgenic mice over expressing RanBP9 in the brain as a part of an ongoing NIH R03 grant. By crossing B6C3-Tg85Dbo mice (APdE9) carrying APPswe, PSEN1dE9 mutations with RanBP9 transgenic mice, RanBP9/APdE9 triple transgenic mice were generated, which produced more CHAPSO-soluble AB&#61472;and c-terminal fragments (CTFs) compared to APdE9 mice as early as 3 months of age, suggesting that RanBP9 increases amyloidogenic processing of APP in vivo. This R01 proposal is an extension of the R03 project. As loss of synapses is a better correlate of the extent of cognitive deficits in Alzheimer[unreadable]s patients and since RanBP9 is present in substantial amounts in neurites and is a strong inhibitor of neurite outgrowth, we next want to examine in this proposal, whether RanBP9-induced altered processing of APP also leads to dendritic and spine injury. We have successfully produced RanBP9 transgenic mice as well as heterozygous null mice for the first time. We propose to compare the pattern of dendritic arborization, spine density, presynaptic and postsynaptic protein levels in the hippocampus and frontal cortex followed by tests for learning and memory skills at 2, 5 and 10 months of age in eight groups of mice, i.e., RanBP9-629 single transgenic, APdE9 double transgenic, RanBP9-629/APdE9 triple transgenic, RanBP9-599 single transgenic, RanBP9-599/APdE9 triple transgenic, RanBP9-/- or RanBP9+/-, RanBP9-/- or RanBP9+/-/APdE9 and wild type litter-mate controls. Neuron Studio, software for automated spine density analysis, will be used to analyze dendritic branching points and spine numbers in Lucifer-yellow-stained pyramidal neurons after obtaining images by laser scanning confocal microscope. If RanBP9 is confirmed as a bona fide target in vivo in this study, the triple transgenic mice may prove to be useful as an accelerated model for synaptic and behavioral deficits.